Intel Xeon E3-1231 v3 vs Intel Xeon E5-2620 v3 Comparison

The Intel Xeon E5-2620 v3 and the Intel Xeon E3-1231 v3 are both end-of-life Haswell-era server/workstation processors, but they target distinctly different platforms. The E5-2620 v3 is a 6-core, 12-thread part built for the Intel Socket 2011-3 platform with quad-channel DDR4 memory, while the E3-1231 v3 is a 4-core, 8-thread part for the Intel Socket 1150 platform with dual-channel DDR3 memory. In head-to-head Cinebench testing, the E5-2620 v3 wins all six benchmarks, with a consistent performance advantage ranging from 10.5% to 10.8% across both multi-core and single-core tests. However, the two processors land at the same 43rd percentile among all CPUs, and their average benchmark scores are remarkably close, differing by less than 1%.

Where Each One Wins

The benchmark data is unambiguous: the Intel Xeon E5-2620 v3 wins every single head-to-head comparison, taking all six Cinebench tests. Its largest margins come in single-core workloads, where it leads by 10.7% in Cinebench R15 single-core (93 vs 84) and 10.8% in Cinebench R20 single-core (391 vs 353). The multi-core victories are nearly as consistent, with the E5-2620 v3 holding a 10.5% advantage in Cinebench R15 multicore (664 vs 601), Cinebench R20 multicore (2769 vs 2506), and Cinebench R23 multicore (6595 vs 5967). These margins reflect the E5-2620 v3’s two additional physical cores and four additional threads, which provide a structural advantage in heavily threaded rendering tasks.

The E3-1231 v3, despite losing every head-to-head test, does not fall far behind. Its 4-core/8-thread configuration still delivers competitive multi-core scores, trailing by roughly 10% in all three Cinebench multi-core tests. Its single-core performance is also within 11% of the E5-2620 v3, which is notable given that the E3-1231 v3 has a higher base clock (3.40 GHz vs 2.40 GHz) and boost clock (3.80 GHz vs 3.20 GHz). The E5-2620 v3’s architectural advantage in IPC and its larger 15 MB shared L3 cache (vs 8 MB) appear to offset the E3-1231 v3’s clock speed advantage, allowing the 6-core part to win even in single-threaded tests where raw frequency usually matters most.

The Verdict

From the data alone, the Intel Xeon E5-2620 v3 is the clear performance winner. It outscores the E3-1231 v3 in every benchmark category, with margins between 10.5% and 10.8% in both single-core and multi-core workloads. If the priority is maximum Cinebench rendering performance, the E5-2620 v3 is the only choice between these two parts. Its 6-core/12-thread design, combined with a larger 15 MB L3 cache and quad-channel memory support, gives it a decisive edge that the E3-1231 v3 cannot overcome despite its higher clock speeds.

However, the E3-1231 v3 is not without its own merits. It achieves slightly better average benchmark scores when considering its full benchmark suite, including Geekbench results not available for the E5-2620 v3. The E3-1231 v3’s average benchmark score is 1925, compared to the E5-2620 v3’s 1907, a difference of less than 1%. This suggests that in workloads not covered by Cinebench, the E3-1231 v3 may hold its own or even pull ahead. The E3-1231 v3 also operates on a more common Intel Socket 1150 platform with dual-channel DDR3 memory, which may offer broader compatibility with existing systems. For users with a Socket 1150 board who need a drop-in upgrade, the E3-1231 v3 is the practical choice. For users building a new system on Socket 2011-3, the E5-2620 v3 provides superior rendering performance and future-proofing with DDR4 memory support.

Head-to-Head Benchmarks

The largest single-core win for the E5-2620 v3 comes in Cinebench R20 single-core, where it scores 391 against the E3-1231 v3’s 353, a 10.8% advantage. This is closely followed by Cinebench R15 single-core, where the E5-2620 v3 scores 93 versus 84, a 10.7% lead. In Cinebench R23 single-core, the E5-2620 v3 scores 931 versus 842, a 10.6% margin. These consistent single-core victories are surprising given the E3-1231 v3’s 1.0 GHz higher base clock and 0.6 GHz higher boost clock. The data suggests that the E5-2620 v3’s Haswell-EP architecture delivers higher instructions per clock than the Haswell-WS architecture in the E3-1231 v3, or that the larger L3 cache plays a significant role in single-threaded performance.

In multi-core workloads, the E5-2620 v3 maintains a steady 10.5% lead across all three Cinebench versions. In Cinebench R15 multicore, it scores 664 versus 601; in Cinebench R20 multicore, 2769 versus 2506; and in Cinebench R23 multicore, 6595 versus 5967. The consistency of these margins across different Cinebench versions indicates that the performance gap is stable and not workload-specific. The E5-2620 v3’s two extra cores and four extra threads provide a theoretical 50% more parallel processing capacity, but the actual performance advantage is capped at 10.5%, suggesting that memory bandwidth or other system bottlenecks limit scaling. The E3-1231 v3’s dual-channel DDR3 memory bandwidth of 25.6 GB/s is less than half the E5-2620 v3’s quad-channel DDR4 bandwidth of 59.7 GB/s, which may explain why the 6-core part does not achieve a larger multi-core lead.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Xeon E5-2620 v3 has 6 cores and 12 threads, while the Intel Xeon E3-1231 v3 has 4 cores and 8 threads.

Q: What is the performance difference in multi-core Cinebench R23?

A: The E5-2620 v3 scores 6595, which is 10.5% higher than the E3-1231 v3’s score of 5967.

Q: Does the E3-1231 v3 win any head-to-head benchmarks?

A: No. The head-to-head data shows the E5-2620 v3 winning all six benchmarks, with the E3-1231 v3 winning zero.

Q: How do their average benchmark scores compare?

A: The E3-1231 v3 has a slightly higher average benchmark score of 1925, compared to the E5-2620 v3’s 1907, a difference of less than 1%.

Q: What memory types do they support?

A: The E5-2620 v3 supports DDR4 memory with a quad-channel bus and 59.7 GB/s bandwidth. The E3-1231 v3 supports DDR3 memory with a dual-channel bus and 25.6 GB/s bandwidth.

Q: What is the difference in their L3 cache sizes?

A: The E5-2620 v3 has 15 MB of shared L3 cache, while the E3-1231 v3 has 8 MB of shared L3 cache.

Architecture Differences

The two processors share the same 22 nm process node and Intel Haswell architecture, but they diverge significantly in their implementation. The E5-2620 v3 uses the Haswell-EP codename and is built for the Intel Socket 2011-3 platform, while the E3-1231 v3 uses the Haswell-WS codename and targets the Intel Socket 1150 platform. This platform difference drives several key architectural distinctions. The E5-2620 v3 features 6 cores and 12 threads with a base clock of 2.40 GHz and a boost clock of 3.20 GHz, while the E3-1231 v3 offers 4 cores and 8 threads with a base clock of 3.40 GHz and a boost clock of 3.80 GHz. The E5-2620 v3’s die size is 356 mm² with 2,600 million transistors, compared to the E3-1231 v3’s 160 mm² die with 1,400 million transistors. This larger die houses a 15 MB shared L3 cache, versus 8 MB on the E3-1231 v3.

Memory architecture is another major differentiator. The E5-2620 v3 supports DDR4 memory over a quad-channel bus, providing 59.7 GB/s of memory bandwidth. The E3-1231 v3 supports DDR3 memory over a dual-channel bus, providing only 25.6 GB/s. This more than 2x memory bandwidth advantage for the E5-2620 v3 is crucial for multi-core workloads that saturate memory bandwidth. The PCIe configurations also differ: the E5-2620 v3 provides 40 PCIe Gen 3 lanes from the CPU, while the E3-1231 v3 offers only 16 PCIe Gen 3 lanes. Both processors support ECC memory and lack integrated graphics, and both are multiplier-unlocked. The E5-2620 v3 has a TDP of 85 watts, slightly higher than the E3-1231 v3’s 80 watts. The E5-2620 v3 was released on September 7, 2014, with a launch MSRP of $417, while the E3-1231 v3 was released on May 10, 2014, with a launch MSRP of $250.